Diffusion Magnetic Resonance Imaging (MRI) is a non-invasive technique for assessing white matter microstructure and long range neuronal connectivity in the brain. It is well known that diffusion tensor imaging (DTI) is sensitive to changes in tissue structure in a wide range of neurological and psychiatric diseases, in many cases detecting changes that are not visible with other imaging methods. In addition, DTI tractography has been used to reconstruct the pathways of major fiber bundles in the living brain and often shows excellent qualitative agreement with expected white matter anatomy, based on dissections of the human brain. While these applications of diffusion MRI are promising, there have been few studies that relate diffusion MRI measurements to the cellular properties they putatively represent. Currently, diffusion anisotropy is thought to reflect axon 'integrity'and/or 'coherence'in white matter. These concepts have no quantitative definition and their relation to actual axon properties have not been established. This project aims to test the correspondence between diffusion MRI and the distribution of myelinated fibers in the primate brain. Specifically, aim 1 involves developing improved methods for comparing axon fiber orientation measurements from diffusion MRI and light microscopy. In addition, the accuracy of spatial registration of the MRI and microscopy data will be quantified. In aim 2, the principal diffusivities and directions obtained with DTI will be compared to the angular distribution of axon fibers determined from myelin stained tissue sections. Further, MR fiber tracking will be compared to the gold standard of fiber tracts defined using neuroanatomical tracers. These tests will be performed in several of the major motor pathways (e.g., commisural, association, and projection pathways). As a specific application, the ability of DTI to localize and characterize the subthalamic nucleus and surrounding tissues will be tested. The results will inform current efforts to use DTI in the planning, placement, and programing of deep brain stimulators in humans. In aim 3, the potential advantages of high angular resolution diffusion imaging (HARDI) will be tested by similar comparisons to the true axon orientation distributions and fiber tracts in motor pathways (determined by light microscopy). In summary, this project will test common assumptions about the information provided by diffusion tensor imaging and high angular resolution diffusion imaging, and therefore provide a better basis for intrepreting and improving measurements made with these techniques.